Main Menu
Welcome
meeting posters and abstracts [ posters presented at meetings and their abstracts ]
| Order in Spontaneous Behaviour | |||
| Author | Bj�rn Brembs, Alexander Maye and Uwe Greggers | ||
| Author email | bjoern©brembs.net | ||
| Author website | http://brembs.net | ||
| Description | Poster presented at the 2005 Neuroscience meeting in Washington, DC. Abstract: It is commonly believed that animals in general and insects in particular are mere input/output machines: if one only knew all their sensory input, one could predict the behavioral output they would produce. This basic tenet not only guides basic neurobiological research but has been the foundation for a great many robotic applications. Our results contradict this view and instead suggest that the brain spontaneously initiates behavioral activity at the same time it is computing input in order to generate output. Our mathematical evaluation detected an ordered but probabilistic temporal structure in spontaneous behavior, indicating a built-in, evolutionarily conserved spontaneity generator in the brain. We hypothesize that this generator can function independently of environmental input and that it evolved to generate flexible behavior in a complex world. Since insects are so commonly used as a metaphor for the robot-like behavior of animals, we used tethered fruit flies ( Drosophila) and foraging honeybees for our study. The flies fly stationarily, attached by head and thorax to a device (torque meter) which measures their tendency to make left or right turns (yaw torque). The flies� environment was made entirely featureless for any of the fly�s senses. Hence, any behavior produced by the flies must be spontaneous and the behavior�s temporal structure will tell us something about the underlying structure of the generator in the brain which produced it. According to the robot-hypothesis, any variability in behavior without environmental input should reflect random noise, much like the hiss of static from a radio tuned between stations. Similarly, we studied the flight paths of bees searching for their hive after displacement to an area lacking natural landmarks. Bees were tagged with harmonic radar transponders and their positions recorded while in flight. These bees were searching for the hive after they had been displaced by 200 m in an area which lacked natural landmarks. The analyses included Geometric Random Inner Products (GRIP), distribution analysis computing L�vy exponents, the computation of correlation dimensions and nonlinear forecasting (simplex-projection and S-map procedure). The idea behind the experiments was that, even in insects, any ordered structure of such spontaneous behavior would have a significant impact on our understanding of basic function of all brains. Previous results have implicated deterministic chaos in certain behavior patterns of freely-moving animals. Systems exhibiting deterministic chaos appear to be random, even though the model of the system is deterministic in the sense that it is well-defined and does not contains any random parameters. Such systems are thus orderly in some sense; this technical use of the word chaos is at odds with common parlance, which suggests complete disorder. The distinction between random noise and deterministic chaos is difficult but important, because the former points to extrinsic, uncontrollable origins of variability, whereas the latter indicates intrinsic origins. Reducing measurement error and other sources of noise will lead to a significant increase in predictability for a brain where the main source of variability stems from noise. In contrast, noise reductions will only marginally change the variability of the output of a chaotic brain whose output is fundamentally indeterministic, despite the deterministic rules that govern it. Our analysis shows that random noise cannot account for the temporal structure of the behavior we analyzed. Instead, the results suggest an underlying mechanism for spontaneous behavior initiation, which evolved to generate probabilistic behavior patterns. Such patterns have been previously shown to outcompete random and deterministic patterns in a number of ecological situations. In the real world, predator avoidance and prey catching behavior spring to mind as other obvious beneficiaries from indeterminacy. One can easily conceive how �getting out of a rut� would also be difficult with only pre-programmed �responses� � �thinking out of the box�, creativity and freedom are required, generated entirely from within; in other words: spontaneity. Future research in this area should include an application of our analysis in a range of other animal and human data and a genetic disruption of targeted fly-brain areas to look for the biological substrate underlying the spontaneity generator. | ||
| Image | no image available | ||
| Filesize | 1.55 MB | ||
| Downloads | 4506 | ||
| Download |  | ||
| Rating |
| ||
You must be logged in to make comments on this site - please log in, or if you are not registered click here to signup